Cl⁻ Secretion in Colonic Epithelial Cells Induced by the Vibrio parahaemolyticus Hemolytic Toxin Related to Thermostable Direct Hemolysin

Abstract

A hemolytic toxin related to thermostable direct hemolysin (TDH), TDH-related hemolysin (TRH), produced by Kanagawa-phenomenon-negative Vibrio parahaemolyticus is suspected of playing an important, but yet-to-be-elucidated role in diarrhea caused by this organism. In cultured human colonic epithelial cells, TRH increases Cl⁻ secretion, followed by elevation of intracellular calcium.

Vibrio parahaemolyticus is an important cause of gastroenteritis, with outbreaks now occurring all over the world (2, 17). However, the mechanisms underlying the diarrheic action of this pathogen are not completely understood. V. parahaemolyticus can be categorized according to the presence or absence of the Kanagawa phenomenon (KP), in which there is hemolysis due to thermostable direct hemolysin (TDH) produced by the organism on Wagatsuma agar medium (14). TDH has been considered a major virulence factor for this disease (8, 23); however, some cases of V. parahaemolyticus gastroenteritis are due to KP-negative strains (10, 15, 16). From clinically isolated KP-negative V. parahaemolyticus, we identified a new hemolysin, named TDH-related hemolysin (TRH), and demonstrated that this TRH was also an important virulence factor (10, 24). For example, TRH stimulated fluid secretion in the rabbit loop test (8, 10). Thus, there is a possibility that TRH induces diarrhea. TRH is immunologically related, but not identical, to TDH ordinary human virulence factor. Unlike that of TDH, the activity of TRH is labile to heat treatment at 60°C for 10 min (10). Although we recently demonstrated that TDH induced Cl⁻ secretion in human colonic epithelial cells (22), there is no evidence how TRH induces fluid secretion.

The present study was designed to investigate how TRH affects short-circuit current (Isc) in human colonic epithelial cell (Caco-2 cell) monolayers and the intracellular Ca²⁺ concentration ([Ca²⁺]_in) in Caco-2 cells after exposure to TRH. The measurements of Isc and [Ca²⁺]_in in Caco-2 cells were taken by following a previously described method (22). Briefly, stock cells were trypsinized, suspended at 20 × 10⁴ cells/ml in medium, and seeded at confluent density onto 1.0-cm² transwell inserts (Costar, Cambridge, Mass.). After 3 days, 2 mM sodium butyrate (Sigma) was added. Sodium butyrate is known to induce differentiation in many cells, including Caco-2 cells (1, 12, 13). It occurs naturally in normal human colon and is used as an energy source by colonocytes (18). Fecal levels of butyrate in healthy humans may be as high as 20 mM (4). Caco-2 cells treated with butyrate have higher transepithelial resistances and greater sensitivity to transepithelial Isc increases by TRH compared with those of nontreated cells (A. Takahashi and T. Honda, unpublished observations). Thus, butyrate-treated Caco-2 cells were used in this study to investigate the effects of TRH. The cells were cultured on the transwell inserts for 6 days and were mounted into a modified Ussing chamber and maintained at 37°C in modified Ringer solution (pH 7.4; gassed with 5% CO₂–95% O₂). Transepithelial resistance was measured by applying a 5-mV pulse at intervals of 40 to 50 s, and the resistance was calculated according to Ohm's law. [Ca²⁺]_in was determined by microfluorometry using a fluorescent dye, 1-(2-(5′-carboxyoxazol-2′-yl)- 6-aminobenzofuran-5-oxy)-2(2′amino-5′-methylphenoxy) ethane-N,N,N′,N′-tetraacetic acid, pentaacetoxy methyl ester (Fura-2/AM; Molecular Probes, Eugene, Ore.). Using a previously described method (22, 23), TRH was purified from a clinically isolated strain of V. parahaemolyticus (RIMD 2210531), which was KP negative (TDH negative) and yielded a single band by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (Fig. 1A), suggesting that the toxins were purified to homogeneity.

FIG. 1.

(A) Sodium dodecyl sulfate-polyacrylamide gel electrophoresis gel of purified preparation of TRH. Lane 1, molecular mass markers (New England Biolabs Inc.) (175, 83, 62, 47.5, 32.5, 25, 16.5, and 6.5 kDa); lane 2, TRH. (B) Effect of gluconate ion and Ca²⁺ depletion on Isc in Caco-2 cell monolayers. ●, Control cells (simple exposure to TRH); ■, Cl⁻ replaced by gluconate ion in bath solution of Ussing chamber; ▴, CaCl₂ was omitted from bath solution and EGTA (1 mM) was added to bath solution of Ussing chamber. TRH was added on the apical side of the Caco-2 cells at the time indicated. ▾, negative control (no exposure to TRH). Values are expressed as means ± standard deviations (n = 5). ∗, Significant difference (P < 0.05) versus negative control.

TRH increased Isc in Caco-2 cell monolayers. Isc increased in Caco-2 cell monolayers when TRH was added to the apical side of the monolayers. When the Cl⁻ in the bath solution was replaced with gluconate ion in both the apical and basolateral sides of the cells, Isc did not increase by addition of TRH (Fig. 1B). This indicates that the influence of TRH on Isc is dependent on extracellular Cl⁻. To deplete Ca²⁺ from the apical cell surface, CaCl₂ was omitted from the bath solution and EGTA (1 mM) was added; in these conditions, after exposure to TRH, there was no change in Isc (Fig. 1B). This indicates that Isc change is also dependent on extracellular Ca²⁺. From these results, we hypothesized that Isc increases after exposure to TRH flow along Ca²⁺-activated Cl⁻ channels.

To explore this hypothesis, we used four kinds of channel inhibitors. Figure 2A shows that the influence of TRH on Isc was inhibited by 4,4′-diisothiocyanatostilbene-2,2′-disulfonic acid (DIDS) (300 μM), an inhibitor of the Ca²⁺-activated Cl⁻ channel (5, 11).

FIG. 2.

Effect of inhibitors on Isc in Caco-2 cell monolayers. TRH (10 μg/ml) was added on the apical side at the time indicated. (A) DIDS (100 μM) was added on the apical side at the time indicated. (B) Effect of Cl⁻ channel inhibitors, DIDS (100 μM), glybenclamide (300 μM), NPPB (100 μM), and gadolinium ion (Gd³⁺) (500 μM). Isc was measured after a 15-min exposure to TRH or 10 min after addition of Cl⁻ channel inhibitor. Values are expressed as means ± standard deviations (n = 5). ∗, Significant difference (P < 0.05 versus TRH.

The cystic fibrosis transmembrane conductance regulator (CFTR) is one of the major Cl⁻ channels in Caco-2 cells, and it is one of the most important Cl⁻ secretion pathways involved in human diarrhea (6). We eliminated the possibility that CFTR is a target for TRH by testing with glybenclamide (300 μM) and 5-nitro-2-(3-phenylpropylamino)benzoic acid (NPPB) (100 μM), both of which are known to be inhibitors of CFTR (5, 19, 21); neither had any effect on the influence of TRH on Isc currents flowing through the Cl⁻ channel (Fig. 2B).

We noted cell swelling within 15 min of adding TRH to Caco-2 cells (unpublished observation), which suggested the possibility that the stretch-activated channels may have opened. We tested with gadolinium ion (Gd³⁺), an inhibitor of stretch-activated channels (3, 7), and found it had no effect on Isc (Fig. 2B). This tends to rule out the association of stretch-activated channels with the Cl⁻ secretion induced by TRH.

These observations are consistent with the conclusion that the effects of TRH on Isc are related to Ca²⁺-activated Cl⁻ channels (20).

Intracellular Ca²⁺ concentration.

[Ca²⁺]_in of Caco-2 cells increased in the presence of TRH (Fig. 3A). After depletion of Ca²⁺ from the cell culture medium by withholding CaCl₂ from the bath solution and EGTA (1 mM), however, [Ca²⁺]_in did not increase. Thus, the increase in [Ca²⁺]_in was due to an influx of Ca²⁺ from the extracellular medium after exposure to TRH.

FIG. 3.

[Ca²⁺]_in change after exposure of Caco-2 cell monolayers to TRH. (A) TRH (10 μg/ml) was added at the time indicated. ●, Positive control (simple exposure to Vp-TRH); ■, CaCl₂ was depleted, and 1 mM EGTA was added to bath solution in fluorometer chamber, followed by addition of TRH and the inhibitors indicated; ▴, negative control (no exposure to Vp-TRH). (B) DIDS (300 μM) and TRH (10 μg/ml) were added at the times indicated. Values are expressed as means ± standard deviations (n = 10).

Next, we checked the effect of DIDS on [Ca²⁺]_in. DIDS did not influence the increasing of [Ca²⁺]_in induced by TRH (Fig. 3B).

Our data indicate that exposure to TRH resulted in greater Isc currents and that these increased currents flowed through the Cl⁻ channels (Fig. 1B and 2).

TRH is a pore-forming toxin in erythrocytes (8, 9), suggesting the possibility that Cl⁻ moves through pores formed by TRH. Our evidence, however, does not support this explanation: DIDS and depletion of Ca²⁺ from extracellular solution inhibited the effect of TRH on Cl⁻ currents (Fig. 1A and 2). This would not have been the case if Cl⁻ was simply passing through TRH-formed pores.

Considered overall, our findings suggest that the Cl⁻ currents induced by TRH are mediated by the Ca²⁺-activated Cl⁻ channel, because (i) there is evidence of [Ca²⁺]_in dependency and (ii) DIDS, an inhibitor of the Ca²⁺-activated Cl⁻ channel, inhibited the Isc (Fig. 1B and 2). Moreover, DIDS decreased the Cl⁻ current without reducing the [Ca²⁺]_in elevation (Fig. 3B), leading to the conclusion that increased Cl⁻ currents were a secondary effect of increased [Ca²⁺]_in.

TDH increased [Ca²⁺]_in and activated the Ca²⁺-activated Cl⁻ channel, resulting in Cl⁻ secretion from the basolateral to the apical cell side (22). The effects of TRH to the Cl⁻ secretion are similar to those of TDH. These confirm that TRH is also an important virulence factor. One should pay attention to TRH-producing V. parahaemolyticus, even if it is a KP-negative strain.